Wcdma network-planning-and-optimization

Click here to load reader

  • date post

    22-Mar-2017
  • Category

    Technology

  • view

    9
  • download

    1

Embed Size (px)

Transcript of Wcdma network-planning-and-optimization

  • ESG Engineering Services Group

    WCDMA Network Planning and Optimization

    80-W0853-1

    Revision B

    May, 2006

  • ii 80-W0853-1 Rev B

    QUALCOMM Incorporated 5775 Morehouse Drive San Diego, CA 92121-1714 U.S.A.

    This technology is controlled by the United States Government.

    Diversion contrary to U.S. law prohibited.

    WCDMA Network planning and optimization

    80-W0853-1

    Revision B

    May, 2006

    Copyright 2006 QUALCOMM Incorporated. All rights reserved

  • 80-W0853-1 Rev B iii

    T A B L E O F C O N T E N T S

    1 Introduction..................................................................................................1 2 WCDMA Network Planning.........................................................................3 3 WCDMA/UMTS optimization methodology.............................................11 4 Conclusion.................................................................................................17 5 References .................................................................................................19

  • iv 80-W0853-1 Rev B

    This page left intentionally blank

  • 80-W0853-1 Rev B 1

    1 Introduction

    The Engineering Service Group at QUALCOMM is working with operators throughout the world to find solutions to the same four challenges that are faced repeatedly: 1) sub-optimal RF optimization, 2) difficulty to tune all the parameters, 3) increasing the reliability of inter-system transition, and 4) providing better in-building coverage. These are complex issues, but solving them can be simplified if a proper deployment process is followed as illustrated in Figure 1, [1], [5]. This process follows a divide and conquer approach, focusing on a selected variable at each step.

    Define network objectives

    Select (re-use)sites

    Verify coverage and capacity

    objectives

    Deploy all network nodes

    RF optimizationService

    Optimization (AMR)

    Serviceoptimization (PS and CS)

    Commercial launch

    Continuous optimization

    Pre-optimization verification

    Network Planning

    Initial Optimization

    Continuous Optimization

    Figure 1: Summary of Network Deployment Steps

    From the different phases of network deployment, Radio (RF) network planning and RF optimization are seen as critical as having long term impacts on both performance and capacity [6]. Proper RF configuration is so important that it even impact the coverage continuity (i.e., inter-system change to GSM [7]) and network evolution, notably with the introduction of High Speed Downlink Packet Access (HSDPA) [4].

  • 2 80-W0853-1 Rev B

    This page left intentionally blank

  • 80-W0853-1 Rev B 3

    2 WCDMA Network Planning

    Ensuring that the RF coverage is sufficient within the network should start naturally with a good RF network plan [2]. Network planning is, may be even more for existing operator, an area that is too often overseen. Greenfield operator have the advantage of starting from a clean plate while incumbent operator already have a set of site that they are compelled to use even if these sites were originally acquire to fit in a network with different coverage and capacity limitation. So what is a good RF network plan? A good network plan should address the coverage and capacity requirement of the area considered, but also be sufficiently flexible to allow network expansion without major change of the existing sites. In WCDMA, the coverage and capacity requirement cannot be considered independently, but should be planned at the same time with proper guidelines. This relation between coverage and capacity is often referred to as the breathing effect of WCDMA. Comparing with TDMA/FDMA technologies, such as GSM, the coverage of a WCDMA network cannot be planned independently of the load on the network. The load on the network will impact the coverage in mainly two different ways, depending on which link (Uplink or Downlink) is considered. On the uplink, as more users are added to the network, higher noise would be detected at the node Bs. This increase in noise, called rise over thermal, requires each of the phones or data cards (UE) to increase its transmit power to overcome this noise increase: effectively the uplink coverage is reduced by this required increase in transmit power. This effect has been documented and can be summarized by the rise-over-thermal versus load curve illustrated in Figure 2: as an example, when the load is 50% of the pole capacity, the coverage is reduced by a factor of 3 dB. On the downlink, the breathing effect cannot be quantified so easily as coverage is impacted by the maximum transmit power assigned to traffic channels and the current load on the network rather than by a quantifiable formula.

  • 4 80-W0853-1 Rev B

    0.0

    2.0

    4.0

    6.0

    8.0

    10.0

    12.0

    14.0

    0% 20% 40% 60% 80% 100% 120%

    Loading [% of pole capacity]

    Ris

    e-O

    ver-T

    herm

    al [d

    B]

    Figure 2: Impact of uplink loading on coverage express by the Rise-Over-Thermal, a.k.a. Interference Margin.

    In any case, predicting coverage is easier, in the early stage of network planning by considering only the pilot channel (CPICH). Once that necessary step is completed, the coverage should be further verified for both links (downlink from Node B to UE and Uplink from UE to Node B) and for all services. For the downlink, CPICH coverage should be verified by considering not only if the received signal code power (RSCP) of the pilot channel (CPICH) is sufficient once all the margins are included, but also by estimating the level of interference generated by the other cells. Such interference is typically quantified by the energy per chip to total received power (Ec/No) of the CPICH. Such quantity effectively estimated how much of the received signal can be used at a given location, or put it in other word, how clean is the signal received. The relation between RSPC and Ec/No is mainly impacted by the loading of the system and the quality of the network plan. This is illustrated in Figure 3 showing the high range of Ec/No for a given RSCP value. It should be noted that the quality of the network plan would be reflected by the number of cells detected at a given location, or to word it differently, the cell overlap: a high quality network plan would be one where a single cell is detected over the majority of the cell area and transition between cells are done over clear boundaries.

  • 80-W0853-1 Rev B 5

    -20

    -18

    -16

    -14

    -12

    -10

    -8

    -6

    -4

    -2

    0

    -120 -110 -100 -90 -80 -70 -60 -50 -40

    RSCP [dBm]

    Ec/N

    o [d

    B]

    Single cell, Low Load 3 Equal cells, High load Figure 3: Relation between Ec/No and RSCP for different cell loading and different network plan quality

    When the loading of the system increases the Ec/No degrades but the RSCP stays constant. Degrading Ec/No is an indication of increased other cell interference which will also increase the need for downlink traffic power (DPCH Ec/Ior, when expressed in relative terms). Power being a limited resource, the higher required transmit power may not be available, thus the coverage not being met in loaded condition: this represent the coverage and capacity trade-off for the downlink in a WCDMA systems. In a similar way, adding sites to provide deeper coverage indoor without controlling the footprint of each of them will increase other cells interference and impact service quality and capacity of the system. It should be noted that the total received signal power (Received Signal Strength Indicator RSSI) is never considered in a WCDMA system as an indication of coverage. It is mainly due to the inability to estimate the quality by this value: 10 weak cells would result in a strong RSSI, but the lack of any dominant server would yield poor system performance. This concept is sometime called pilot pollution, where multiple servers contribute to a high RSSI, but where the signal cannot be used due to lack of strong dominant server. To ensure that these issues will be minimized, several simple steps can be taken as illustrated and detailed below.

  • 6 80-W0853-1 Rev B

    Defining the network requirements (coverage area, coverage depth, expected traffic, traffic models ) is necessary to dimension the network both for coverage and capacity.

    Defining the number of site required for a given coverage depth: the site configuration, antenna height and downtilt notably, should be selected as a function of this number of site. Without selecting the site configuration relative to the site-to-site distance, the risk is to have either insufficient coverage or excessive downlink interference. Unlike in GSM network planning where sub-optimal site configuration can be compensate with frequency planning, the 1/1 frequency re-use of WCDMA does not allow such flexibility.

    Defining up front the number of sites required for capacity over the next few years: this number should be compare to the number of site for coverage to ensure that coverage, short term, and long term capacity needs are met. For capacity limited design, the site configuration should match the higher site count. For capacity limited design, the decision between adding sites and adding carrier should consider the possible site configurations. In particular, adding sites with limited flexibility on the antenna configuration may not always add capacity: if the added sites increase the downlink interference, the capacity of each of the sites will decrease.

    Nominal Design

    Sit